Study objectives: Surgical resection remains the treatment of choice for anatomically resectable non-small cell lung cancer. However, the presence of associated comorbid conditions increases the risk of death and surgical complications. Several studies have evaluated the usefulness of preoperative exercise testing for predicting postoperative morbidity and mortality. The aim of this study was to establish whether exercise testing could predict poor surgical outcome in lung cancer surgery and whether the absolute value or percentage of predicted value is the better predictor of the surgical outcome.
Design: The study was designed as a prospective study.
Patients and setting: One hundred thirty patients with potentially operable lung cancer at Papworth Hospital over 2 years were recruited; of these, 101 underwent curative surgery.
Interventions: Spirometry and cardiopulmonary exercise tests were performed for every patient (n = 99), except for two patients with back problems. We also recorded the outcome of surgery, in particular, complications and mortality.
Measurements and results: Mean maximum oxygen transport at peak exercise (Vo(2)peak) was 18.3 mL/kg/min (SD, 4.7 mL/kg/min), and mean percentage of predicted Vo(2)peak value was 84.4% (SD, 30%). Poor surgical outcome was significantly related to Vo(2)peak percentage of predicted (p < 0.01) but not to the actual oxygen uptake value.
Conclusions: The use of the percentage of predicted Vo(2)peak value would be a better indicator of surgical outcome, since it predicts the surgical outcome better, and corrects for normal physiologic ranges. The threshold of Vo(2)peak for surgical intervention could be set between 50% and 60% of predicted without excess surgical mortality.